Drug combinations that include 2 or more nucleoside reverse transcriptase inhibitors (NRTIs) have become the standard of care therapy for individuals infected with human immunodeficiency virus type 1 (HIV-1) and are given to both adults and pregnant women. Our studies in this area focus on genotoxicity and mitochondrial toxicity in cell culture, primates and humans. Cell Culture. Genotoxic events induced by zidovudine (AZT) were followed in human lymphoblastoid MOLT-3 cells exposed continuously to AZT (800uM) for 14 passages, and grown in the absence of drug for an additional 5 passages. AZT-DNA incorporation levels were similar at passages 1, 4 and 14. Untreated cells were passaged once daily, but for the first 5 passages of AZT exposure cells grew slowly and were passaged only every 3-6 days. During passages 1-5 of AZT exposure, active TK1 enzyme protein, which performs the first step in AZT metabolism, was depleted and stayed depleted for the remainder of the study, even in the absence of AZT. Flow cytometry showed that slow-growing cells with depleted TK1 also had an accumulation of cells in S-phase, and removal of AZT only partially reversed the S-phase arrest. To investigate this S-phase arrest, MOLT-3 cells cultured with 800uM AZT for 14 passages were examined by NCI oligo-microarray, cell cycle-specific super array, and Western blot. At passage 5 there were increases in cyclin A, cyclin D3 and retinoblastoma (Rb) binding protein and micronucleus formation, and proliferating cell nuclear antigen was decreased at both passages 5 and 14. Therefore, the AZT-induced S-phase arrest appears to correlate with depletion of TK1, and to be consistent with changes in expression of various cell cycle-related proteins. These experiments explored chromosomal integrity in Chinese hamster ovary (CHO) and normal human mammary epithelial cells (NHMEC) exposed to AZT, and we found several novel forms of chromosomal damage. For example, centrosomal amplification/fragmentation was observed in both cell types in correlation with dose. NHMEC strain M99005, previously identified to incorporate high levels of AZT into DNA (HI), showed maximal centrosomal amplification when compared with a second strain, NHMEC M98040, which did not incorporate AZT into DNA (LI). Additionally, an abnormal tubulin distribution was observed in AZT-exposed HI cells bearing multiple centrosomes. Immunofluorescent staining of NHMEC with Aurora A, a kinase involved in the maturation of the centrosome, confirmed the induction of abnormal chromosomal distribution. Flow cytometric studies revealed that NHMEC bearing abnormal numbers of centrosomes and abnormal tubulin distribution had cell cycle kinetics similar to cells with normal tubulin, indicating that unbalanced chromosomal segregation or aneuploidy, which is known to play a pivotal role in malignant transformation, likely occurs in the daughter cells. Therefore, AZT induces genomic instability and clastogenicity as well as alterations in proteins involved in centrosomal activation, all of which may contribute to the carcinogenic potency of this drug. Evidence of toxicity associated with clinical use of the NRTIs underscores the need for alternate strategies. We hypothesized that the radioprotective agent amifostine (Ethyol), and its active metabolite WR-1065, might reduce NRTI-induced toxicity, and designed studies to examine the antiretroviral efficacy of AZT in the presence of WR-1065. Phytohemagglutinin (PHA)-stimulated T-cell blasts from healthy donors were infected for 2 hr with HIV-1, and viral replication (p24 protein) was determined in the presence of AZT, WR-1065, or the combination, 72 hr later. Both WR-1065 and amifostine inhibited HIV-1 replication by 80% at 50 M. For WR-1065, cell viability was 70-90% at less than/equal to 26 M, and 50% inhibition of HIV-1 replication was at 17.6 and 1.5 M WR-1065 (n=4). Flow cytometric analysis found no cell cycle alterations at 9.5 and 18.7 M WR-1065, and the presence of less than/equal to 26 M WR-1065 did not reduce AZT antiretroviral efficacy. We cultured CD8+ T-cell-depleted, PHA-stimulated T-cell blasts from 3 Macaca mulatta monkeys (macaques) chronically infected (14 months) with simian immunodeficiency virus (SIV). We found near total inhibition of SIV replication (p27) in cells from 2 macaques that were exposed for 17 days to 9.5 and 18.7 M WR-1065. Cells from a third monkey, with the highest viral load, showed inhibition of SIV replication only at the highest WR-1065 dose. These studies showed that WR-1065 has anti-HIV-1 activity in vitro and anti-SIV activity ex vivo. Therefore, WR-1065 and/or amifostine may be clinically beneficial for inhibiting HIV-1 replication (patent pending). Animal Models. We are performing ongoing investigations of mitochondrial toxicity in fetal Erythrocebus patas monkeys, taken at birth, 1 year and 3 years of age, from dams exposed to human-equivalent protocols containing lamivudine (3TC), AZT, AZT/3TC, AZT/didanosine (ddI), 3TC/stavudine (d4T), AZT/3TC/abacavir (ABC) or AZT/3TC/nevirapine (NVP). To model human clinical use, pregnant patas dams (n=3-4/group) are dosed during the last half of gestation, and the neonates for 6 weeks after birth. Biomarkers include clinical chemistry, electron microscopy (EM) for mitochondrial morphology, scoring of damage in EM photos, quantitation of mitochondrial DNA (mtDNA) using PCR-based hybrid capture-chemiluminescence immunoassay (HC-CIA), oxidative phosphorylation (OXPHOS) enzyme assays, and OXPHOS histochemical staining. Studies that include 1 or 2 NRTIs, with time points at birth and 1 year are complete, and data for the heart and skeletal muscle have been published. These papers have documented substantial mitochondrial morphological abnormalities and alterations of mtDNA levels that do not appear to be reversible by 1 year of age, with milder OXPHOS and cytochemical alterations that appear to improve at 1 year. A paper describing NRTI-induced changes in liver and brain is in preparation, and the data show that the effects are less severe than for heart and skeletal muscle, but these organs are not normal at 1 year in the NRTI-exposed animals. Investigation of patas fetuses exposed to AZT/3TC/NVP (2 NRTIs and a non-NRTI) and AZT/3TC/ABC (3 NRTIs) continues with the same end points of interest. Because our early studies showed that there was considerable persistent damage observed at 1 year of age by morphologic, biochemical and molecular end points, and because we were interested in documenting the duration of these effects, we are maintaining some of the animals up to 3 years of age. EM photos obtained to date in monkeys exposed to AZT/3TC/NVP show damage at birth and at 1 year of age in the heart, skeletal muscle and pancreas, which is similar to that found in fetuses exposed only to the 2-NRTI combinations. The data suggest that the most damaging components of these drug combinations are the NRTIs; however, the addition of NVP did result in delay of fetal development and morphological changes in the heart. We have hypothesized that amifostine and WR1065 might also protect against mitochondrial toxicity induced by NRTI exposures because they are strong antioxidants. This is an important question since it would be optimal to prevent the mi [summary truncated at 7800 characters]